ITS FORMATION. 95 



metamorphosis predominates), comparatively simple substances 

 should unite to produce more complicated ones in the formation 

 of excreted matters. 



If we attribute to the fibrin which is decomposed in the liver 

 a great share in the formation of the above-named adjuncts, we 

 must also at the same time meet the objection which may be 

 brought forward, that the fibrin may be applied to the formation 

 of the young blood-corpuscles which are found in such large num- 

 bers in the hepatic veins. I have certainly never found the 

 characters of portal fibrin to differ so much from that of other 

 venous blood, in newly-killed animals, as has been observed by F. C. 

 Schmid ;* it appears, however, to be less contractile and less dense 

 than that of other venous blood. This, at all events, appears not 

 to be the form in which it can be applied to the construction of 

 tissues or blood-corpuscles. Moreover, we see from a comparison 

 of the portal serum with that of the hepatic veins, that the albu- 

 men in the latter is considerably diminished, and has probably 

 been employed in the formation of blood-cells. According to my 

 investigations, the albumen is to the other solid substances in portal 

 serum as 100 : 12 -5, while in the serum of hepatic venous blood 

 (where, moreover, the salts are diminished by about 0*3) the ratio 

 is as 100:27*4. Moreover, hepatic venous blood contains, both 

 absolutely and relatively, far less serum than portal blood ; when 

 for instance, the intercellular fluid is to the moist blood-cells in the 

 ratio of 100 to 150 (and this was the case when horses were 

 killed five hours after feeding), the corresponding ratio in hepatic 

 venous blood is as 100 : 330; or if (ten hours after feeding) the 

 ratio in portal blood is as 100 : 35, the corresponding ratio in hepatic 

 venous blood is as 100 : 138. Hence the portal blood, in its 

 conversion into hepatic venous blood, loses a very considerable 

 portion of its serum, while the latter parts with much of its albumen. 

 The coagulable, soluble albumen of the portal blood has, therefore, 

 in a great part passed over into the considerably augmented cruor 

 which is formed by the blood of the hepatic veins. If it is not too 

 bold an hypothesis to assume that this portion of the albumen is 

 applied to the formation of the walls of the blood- corpuscles, this 

 readily explains why the bile is so rich in sulphur ; for, as we shall 

 prove in a future page, the walls of the corpuscles of hepatic 

 venous blood contain no sulphur. 



Another important constituent of the bile is the pigment, which 

 also cannot be detected pre-formed in the portal blood ; and we have 



* Op. cit. 



